The delivery of fluids plays an important role in fields such as chemistry, microbiology and biochemistry. These fluids may include liquids or gases and may provide reagents, solvents, reactants, or rinses to chemical or biological processes. While various microfluidic devices and methods, such as microfluidic assays, can provide inexpensive, sensitive and accurate analytical platforms, fluid delivery to the platform can add a level of cost and sophistication, as the operation of microfluidic devices often requires the ability to exchange fluids between the device itself and the outside world. In some cases, the operation of the device includes one or a combination of the following: introduction of a sample, introduction of reagents, extraction of a fluid for off-chip analysis or transfer of fluids from one chip to the next.
Because microfluidic devices benefit from scaling law, most applications require only minute quantities of fluid to carry out assays, compared to their bench-top counterparts. Along with the development of these miniaturized systems, the microfluidic community has invested many efforts in designing interfaces between the microfluidic device and the laboratory world. The major problem associated with world-to-chip connection is the mismatch between the volumes used on-chip (e.g., femtoliters to microliters) with respect to the volumes typically handled at the bench (e.g., microliters to liters). For instance, many world-to-chip connectors have dead volume, e.g., wasted volume that may lie at the core of the connector itself. For example, in the case of a tubing with a small inner diameter (e.g., 200 μm to inject small quantities of fluid) connecting to a microchannel (e.g., 10-200 μm in diameter), there may remain a gap between the edge of the tubing and the entrance of the microchannel. The volume defined by that gap is referred to as dead volume, and, in some instances, can be of the same order of magnitude as the total volume of sample to be analyzed. In practice, the dead volume of many devices can often be higher than the volume of sample analyzed by the chip; this is an undesired effect for applications that rely on small sample/reagent consumption.
Accordingly, advances in the field that could, for example, reduce the dead volume and/or allow easy interface between the microfluidic system and the user would be beneficial.